JP2907706B2 - Voice switch control method used with echo canceller - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a control system for a voice switch used together with an echo canceller having an echo canceling function for a telephone line, and more particularly to a control system for performing a stable switch operation.

[0002]

2. Description of the Related Art At a connection between a two-wire telephone line and a four-wire telephone line, a so-called echo occurs in which a part of a received signal leaks to a transmitting side due to impedance mismatch. A similar echo occurs in an electronic conference system or the like.

As shown in FIG. 2, an echo canceller having a function of suppressing this echo is provided with a shift register 31 for storing a reception input sequence Xj inputted from a reception input terminal 1 and a reception output terminal 2 to a transmission input terminal 4. A memory 32 for storing the estimated value Hj of the impulse response of the echo path leading to the input signal, a convolution calculator 33 for convolving the received input sequence Xj and the estimated value Hj of the echo to generate a pseudo echo, A subtractor 34 for outputting a value obtained by subtracting the pseudo echo to the transmission output terminal 8
And a control unit 37 for correcting the estimated value Hj of the echo, and a multiplier 35 and an adder 36 for performing an operation for this correction.

In this echo canceller, when a reception input xj is input from a reception input terminal 1, it is input to a shift register 31.
As a vector Xj. The reception input xj is output from the reception output terminal 2 at the same time, and a part of the reception input xj is input to the transmission input terminal 4 as an echo Yj.

The transmission input terminal 4 receives a transmission signal nj including ambient noise and the like. Therefore, the transmission input terminal 4
, A signal of sj = Yj + nj is input, and a transmission signal nj
Affects as a disturbance to the echo signal Yj.

On the other hand, an estimated impulse response of an echo path from the reception output terminal 2 to the transmission input terminal 4 is stored as a vector Hj in a memory 32 inside the echo canceller. A convolution operation with the received input sequence Xj is performed to generate a pseudo echo yj.

The subtracter 34 calculates a residual echo ej = sj-yj by subtracting the pseudo echo yj from the transmission input sj,
This is transmitted from the transmission output terminal 8. As a result, the return of the echo is suppressed.

Next, the control unit 37 calculates a square norm value {Xj} ² of the received input sequence, and corrects the estimated value Hj using this value and the residual echo ej output from the subtractor 34. The coefficient α · ej / {Xj} ² is calculated, and the calculation result is output to the multiplier 35. The multiplier 35 multiplies the received input column vector Xj output from the shift register 31 by the correction coefficient and outputs the result to an adder 36. The adder 36 multiplies the estimated impulse column vector Hj by the multiplier 35. And the estimated value Hj stored in the memory 32 is corrected by the added value.

The echo canceller alternately executes an operation for canceling an echo and an operation for correcting an estimated value by an algorithm called a learning identification method, and gradually brings the estimated value closer to a true value.

The operation of canceling the echo and the operation of correcting the estimated value are performed by the following equations (1) and (2). Echo cancellation operation;

(Equation 1) Estimated value correction operation;

(Equation 2) N: number of taps xj: reception input sj: transmission input (= echo yj + disturbance nj) ej: transmission output (= residual echo) hj (m): estimated impulse response (time j, mth sample) α: step gain (0) <Α <2).

When the echo cancellation operation of equation (1) and the estimation value correction operation of equation (2) are alternately performed in accordance with the procedure of the learning identification method, a period (single period) where the level of the transmission signal nj is lower than the level of the echo Yj Talk period), Hj
Converges asymptotically to the true impulse response, increasing the echo suppression effect.

A period during which the level of the reception input xj decreases and the level of the transmission signal nj exceeds the level of the echo Yj is called a double talk state, and it is estimated that the correction operation of the equation (2) is performed during this period. The value is greatly disturbed. for that reason,
The control unit 37 determines the level ratio (Spj / Epj) of the transmission input sj and the transmission output ej to the level ratio of the transmission input and the transmission output measured during single talk (the cancellation amount (Spm / E
pm)), when the former falls below the latter, it is determined that a double talk state has occurred, and the correction calculation of Expression (2) is stopped.

As described above, in the echo canceller, the pseudo echo for canceling the echo component from the transmission output approaches the true echo size with the lapse of time due to the correction of the estimated value. Rise rapidly. However, if there is circuit noise or non-linear distortion in the echo path, the amount of cancellation is limited and a large residual echo remains. Also,
Immediately after turning on the echo canceller power supply or immediately after a change in the echo path, the amount of cancellation is reduced, and the echo cannot be sufficiently suppressed.

For this reason, a voice switch is provided in addition to the echo canceller to further suppress the residual echo ej output from the echo canceller. This voice switch compares the reception input level xj of the reception line with the residual echo level ej of the transmission line output from the echo canceller, and inserts a large loss (attenuation) into the line with the smaller level. The return of the echo is prevented by narrowing the line in one direction.

[0015]

However, in the conventional echo suppression system using both an echo canceller and a voice switch, it takes time to calculate the level of the received input level and the residual echo level and to compare and judge them. Switching of the inserted large loss cannot be performed quickly following the level change, and the sound at the beginning of the speech is suppressed, that is, a so-called break of the beginning of the speech occurs.

Further, when there is a large delay in the echo path, there is a problem in that a change in the received input level and the residual echo level causes a time lag, so that an error occurs in selection of the insertion time of the loss. I have.

In a voice switch used in combination with an echo canceller, it is necessary to turn off insertion of a loss in a double talk state. The detection of the double talk is determined by the level ratio of the transmission input sj and the transmission output ej ( Spj / E
pj) and the amount of cancellation (Spm / Epm), when the conventional method is used, when the amount of cancellation converges to near the upper limit, slight fluctuations in the ambient noise included in the transmission signal nj And the detection result appears differently. For that reason, double talk and single talk are frequently switched,
The voice switch performs an unstable on / off operation in conjunction with the voice switch, and cannot stably suppress the echo.

The present invention solves such a conventional problem, and suppresses the residual echo by a stable switch operation without being affected by the delay of the echo path and without causing the start of speech. It is an object of the present invention to provide a control method of a voice switch that can be used.

[0019]

Therefore, in the present invention, in a control method of a voice switch used together with an echo canceller, a transmission signal level Spj including an echo signal at the time of single talk and an output signal level Epj of the echo canceller are provided.
Is stored as the amount of cancellation (EpM / SpM), the shortage of the predetermined amount of echo suppression with the amount of cancellation is determined as the insertion loss LOSS, and the transmission signal level Spj and the echo canceller of the echo canceller subsequently input. Output signal level E
pj and the amount of cancellation (EpM / Spj)
M) and when 0 <C ₁ <C ₂ <1, (1) Spj / Epj ≦ C ₁ * (SpM / EpM)
In the case of ( ₁ ), the correction of the estimated value in the echo canceller is stopped, and the output of the echo canceller is directly output from the voice switch, and (2) C ₁ * (SpM / EpM) <
If Spj / Epj <C ₂ * (SpM / EpM), stop the correction of the estimated value in the echo canceller,
(3) Spj / Ep is output from the voice switch by adding the insertion loss LOSS to the output of the echo canceller.
When j ≧ C ₂ * (SpM / EpM), the estimated value is corrected in the echo canceller, and the insertion loss LOSS is output from the voice switch to the output of the echo canceller.
Is added and output is performed.

[0020]

In this control method, the voice switch is controlled by selecting data which is not affected by the delay of the echo path. Therefore, there is no possibility that the control is disturbed by the delay of the echo path.

Further, the insertion loss LOSS amount is limited to a minimum amount that needs to compensate for the echo suppression amount of the echo canceller in order to obtain a desired echo suppression amount. Therefore,
Even if the speech signal at the beginning of the conversation is attenuated by the insertion loss, the amount is small, so that the occurrence of the beginning of the talk can be avoided.

Further, the reference point at which the application of the insertion loss LOSS to the output of the echo canceller is stopped is set to a position lower than the reference point at which the echo canceller determines that double talk is performed. Therefore, when the double talk and the single talk are frequently switched due to a slight change in the ambient noise, the voice switch continues to apply the insertion loss LOSS stably to the output of the echo canceller, and complete the double double talk. When the talk state is reached, the insertion loss L
The application of the OSS is stopped.

[0023]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS As shown in FIG. 1, an audio switch implementing a control method of the present invention calculates an insertion loss amount and determines whether to insert the loss.
A variable attenuator 6 for adding the attenuation of the value calculated by the insertion loss calculator 5 to the transmission output of the echo canceller 3; and turning on / off insertion of a loss into the transmission output based on the determination result of the insertion loss calculator 5. And a switch 7 for turning off.

The echo canceller 3 receives the input xj from the input terminal 1 and takes it in, and generates a pseudo echo by the operation described above.

The reception input xj is output from the reception output terminal 2, and a part of the reception input xj is input to the transmission input terminal 4 as an echo Yj. The transmission input terminal 4 has an echo signal Y to be suppressed.
j, a transmission signal nj which should not be suppressed is superimposed and input. As a result, the transmission input sj =
nj + Yj is input.

The echo canceller 3 outputs a residual echo ej obtained by subtracting a pseudo echo from the transmission input sj. At the same time, when the echo canceller 3 is in the single talk state (when the level of the reception input xj is sufficiently large and the level of the echo Yj greatly exceeds the level of the transmission signal nj), the square value of the residual echo ej is minimized. Then, the stored estimated impulse response is sequentially corrected. By this correction, the estimated impulse response converges toward the true value, and accordingly, the power of the residual echo ej decreases, and the echo is suppressed.

The quantity representing the performance of the echo canceller 3 includes a transmission input s in a single talk state.
The “cancellation amount SpM / EpM” that represents the ratio between the level Spj of j and the level Epj of the residual echo ej is defined.
This cancellation amount can be obtained as an average level of a plurality of sample values in the single talk state, or
It can be obtained from the transmission input level of the sample and the residual echo level in the single talk state where the echo is sufficiently suppressed. Further, the cancellation amount SpM / EpM can be updated in a single talk period after a lapse of a predetermined time, if necessary. The subscript M is
The amount is stored in the single talk state and does not change in the double talk state.

The correction of the estimated impulse response is not performed in the double talk state because the estimated value is disturbed by the transmission signal nj in this state. In order to determine the double talk state, the echo canceller 3 checks whether or not the ratio (Spj / Epj) between the transmission input level Spj and the residual echo level Epj at that time satisfies the following equation (3). When this is satisfied, it is determined that the state is the double talk state, and the execution of the correction calculation is stopped.

Spj / Epj <C ₂ * (SpM / EpM), C ₂ = 1/2 (3) On the other hand, the insertion loss calculator 5 receives the transmission input signal sj and the output signal ej of the echo canceller 3. , The transmission input signal level Spj and the output signal level Ep of the echo canceller
j is calculated, and the cancellation amount SpM / EpM representing the ratio of Spj and Epj at the time of single talk is calculated based on these j and is stored.

Next, the insertion loss calculator 5 calculates the insertion loss L
OSS is determined by the following equation (4).

LOSS = L * (SpM / EpM), L = 10 ^{− (50/20)} (4) Here, L represents an echo by the action of both the system, ie, the echo canceller 3 and the voice switch. The desired amount of echo suppression to be suppressed, LOS
S represents an amount that is insufficient to suppress the echo canceller 3 alone to obtain L.

Further, the insertion loss calculator 5 calculates the level ratio (Spj / Epj) of the newly input transmission input signal sj and the output signal ej of the echo canceller 3 and the cancellation amount Sp.
M / EpM is compared by the following equation (5), and based on the comparison, it is determined whether or not the insertion loss LOSS calculated by the equation (4) is actually given.

Spj / Epj <C ₁ * (SpM / EpM), C ₁ = 1/4 (5) When the expression (5) is satisfied, a complete double talk state is established. Insertion loss L
No OSS is given. Conversely, when the equation (5) is not satisfied, an insertion loss LOSS is given.

The switch 7 outputs the output of the echo canceller 3 to the transmission output terminal 8 as it is when the equation (5) is satisfied, based on the determination result of the equation (5) by the insertion loss calculator 5, and If (5) does not hold, the output of the echo canceller 3 attenuated by the variable attenuator 6 is output to the transmission output terminal 8.

Accordingly, the echo canceller 3 and the voice switch provide the ratio (Sp) between the newly input transmission signal level Spj and the output signal level Epj of the echo canceller 3.
The following three-stage operation is performed according to the magnitude of (j / Epj) (provided that 0 <C ₁ <C ₂ <1). (1) When Spj / Epj ≦ C ₁ * (SpM / EpM), complete double talk is performed, and echo canceller 3
Stops the correction of the estimated value, and the voice switch outputs the output of the echo canceller 3 as it is. (2) C ₁ * (SpM / EpM) <Spj / Epj <C ₂
In the case of * (SpM / EpM), the echo canceller 3 determines double talk and stops correcting the estimated value. The voice switch outputs the output of the echo canceller 3 by adding the insertion loss LOSS calculated by the equation (5). (3) When Spj / Epj ≧ C ₂ * (SpM / EpM), the echo canceller 3 determines that the signal is a single talk and corrects the estimated value. The insertion loss LOSS calculated in step (1) is added and output.

As described above, in the control method of the embodiment, the reference point at which the application of the insertion loss LOSS to the output of the echo canceller 3 is stopped is lower than the reference point at which the echo canceller 3 determines that double talk is performed. Stipulated. As a result, when double talk and single talk are frequently switched due to changes in ambient noise, etc.,
The audio switch stably applies the insertion loss LOSS to the output of the echo canceller 3, and stops applying the insertion loss LOSS when the double talk state is completely reached.

In this voice switch, of only two lines connected by the echo path, data of the line on the transmission input terminal 4 side is used to calculate an insertion loss, determine a double talk state, and stop loss insertion. The timing is determined. Therefore, the operation of the voice switch is completely unaffected by the echo path delay.

In this voice switch, the insertion loss L
As the OSS, a minimum necessary attenuation for compensating the echo suppression amount of the echo canceller 3 is set. Therefore, even if the insertion time of the loss is shifted and slightly overlaps with the start time of the conversation, it is possible to avoid a large attenuation of the audio signal, and to avoid occurrence of the beginning of the talk.

[0039]

As is apparent from the above description of the embodiment, in the voice switch control method of the present invention, the switch operation can be performed without being affected by the delay of the echo path.

Further, even when the cancellation amount of the echo canceller has converged to near the upper limit, the voice switch can be operated stably.

Further, it is possible to avoid occurrence of the beginning of a talk, and to prevent deterioration of speech quality.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a configuration of a voice switch that implements a control method according to the present invention;

FIG. 2 is a block diagram showing a configuration of a conventional echo canceller.

[Explanation of symbols]

Reference Signs List 1 reception input terminal 2 reception output terminal 3 echo canceller 31 shift register storing reception position sequence Xj 32 memory storing estimated impulse response Hj of echo path 33 convolution calculator 34 subtractor 35 multiplier 36 adder 37 controller 4 Transmission input terminal 5 Insertion loss calculator 6 Variable attenuation unit 7 Switch 8 Transmission output terminal

Claims (1)

(57) [Claims] 1. A control method of a voice switch used together with an echo canceller, wherein a transmission signal level S including an echo signal at the time of single talk is provided.
The ratio between pj and the output signal level Epj of the echo canceller is stored as the amount of cancellation (EpM / SpM), and the shortage of the predetermined amount of echo suppression relative to the amount of cancellation is obtained as the insertion loss LOSS. The ratio (Spj / Ep) of the transmission signal level Spj to be output to the output signal level Epj of the echo canceller
j) and the cancellation amount (EpM / SpM),
If the relationship of <C ₁ <C ₂ <1, and if Spj / Epj ≦ C ₁ * (SpM / EpM), the correction of the estimated value in the echo canceller is stopped, and the echo switch is switched from the voice switch to the echo canceller. Is output as it is, and C ₁ * (SpM / EpM) <Spj / Epj <C ₂ * (S
In the case of (pM / EpM), the correction of the estimated value in the echo canceller is stopped, and the output of the echo canceller is added with the insertion loss LOSS from the voice switch to be output. Spj / Epj ≧ C ₂ * ( In the case of (SpM / EpM), the echo canceller corrects an estimated value, and a voice switch outputs the echo canceller with the insertion loss LOSS added thereto. .